John K Clark (johnkc@well.com)
Thu, 12 Jun 1997 20:53:50 -0700 (PDT)


Max More <maxmore@primenet.com> wrote:

>It's not obvious that a change of platform will instantly make us
>totally malleable or give us the ability easily to simply plug in
>new abilities.

I agree, sort of. If you just changed the hardware platform, such as
replacing biological neurons with nanotechnology ones, it would not give you
new abilities, except for much greater speed and memory. You could recite
from memory any page from a book on advanced Physics, but you might not
understand what it meant.

The great thing about a well designed computer platform is that once somebody
writes a program that gives a machine a new ability it's very easy to give
that ability to all the other machines on the platform. The trouble with the
Human brain is that it is not well designed, it is not designed at all, it's
the product of millions of years of random mutation and natural selection
producing just what you would expect, hideous spaghetti code. Well designed
programs are easy to modify, with a sloppy program it's hard to even figure
out how the damn thing works much less modify it, that's why it takes years
to learn to play the piano. A master pianist can't just transmit his skill to
others in a few seconds, so we must continually reinvent the wheel.

I'll bet one of the first things an upload would try to do is untangle this
messy code of his into a simpler more logical structure and then recompile
it. This would be a gargantuan task that could take us thousands of years,
even an upload might be at it for days before he was successful.

On Thu, 12 Jun 1997 Hagbard Celine <hagbard@ix.netcom.com> Wrote:

>it seems unlikely to me (stick neck out) that an ability arises out
>of a mere confluence of data. Skills like playing the piano don't
>seem to arise out of your knowledge of music since an eye-hand motor
>skill is involved.

In theory all skills must be represented by data. If we have skills then our
neurons must act that way, and they act that way if they have the correct
structure, and they have the correct structure if the atoms in them are in
the right place, and data will tell you if they are in the right place.
Of course, figuring out what placement of atoms will let you play the piano
is very far from trivial.

On Thu, 12 Jun 1997 Hal Finney <hal@rain.org> Wrote:

>One good news/bad news aspect of uploading is that it should be
>possible to run the simulation a lot faster than an organic brain.


>This is too bad, because actually in most ways the quality of life
>will be worse for the super-fast brain. From the brain's point of
>view, it is as though the rest of the world is slowed down by a huge
>factor. Everything will be slow - computers, communication,

If technology had advanced enough so we could upload a human mind then we
would certainly have computers and communication links enormously faster than
anything we have now, fast enough to keep up with any upload.

>any manipulation of the physical world. If you need to do experiments
>to learn more about nature, they will take (subjective) centuries to

The environment operates at all levels from Pico seconds to billions of years,
all equally valid, the time frame of a biological human brain does not occupy
a preferred position in the universe. For example, the arm on a Nanotechnology
based Assembler would be 50 million times shorter than a human arm, and that
means it could move back and forth 50 million times faster than a human arm.
Also, just one pound of carbon has enough atoms to make a billion billion
such Nano Robots. I think even the fastest upload would find plenty to do.

>If you expect to live in a nifty VR, think again: computers will be
>a million times slower (subjectively) which will limit the quality
>of the simulations they can provide. You may have to live in a
>cartoon world.

Concerning Virtual Reality, I don't see how speed could even be an issue for
an upload. Even if the computer that was emulating you and the virtual world
was very slow, from your point of view it would seem infinitely fast. If the
machine had performance problems all you'd have to do is have the part of the
program that was simulating you slowed down or even stopped, while leaving
the part of the program that simulated the rest of the universe running at
normal speed. Regardless of how many calculations it would take to convince
you that the simulation was real it could all be done instantly, from your
point of view. Once the machine was caught up, your part of the program could
be carefully restarted till the next speed bottleneck.

John K Clark johnkc@well.com

Version: 2.6.i